Radiology AI Directory

FDA-cleared profile

Auto Segmentation

GE Medical Systems, LLC

Auto Segmentation generates initial contours for 40 adult organs or subregions from contrast or noncontrast CT across the head and neck, thorax, abdomen, and pelvis.

Radiotherapy planningQuantification and segmentation
Evidence status: each field shows source quality, applicability, and review date. Research pending, information not established in reviewed sources, and vendor documentation pending are distinct outcomes.

Clinical fit

What this tool is for

Start with the authorized purpose, then verify how it fits your service line and reading workflow.

Exact purpose
Auto Segmentation generates initial contours for 40 adult organs or subregions from contrast or noncontrast CT across the head and neck, thorax, abdomen, and pelvis.
FDA source Exact FDA submission Checked 2026-09-03
Intended users
Dosimetrists, medical physicists, radiation oncologists, and qualified radiotherapy practitioners.
FDA source Exact FDA submission Checked 2026-09-03
Care setting and population
Radiation-oncology planning departments using compatible CT and RT Structure Set review systems.
FDA source Exact FDA submission Checked 2026-09-03
Workflow role
Automatic initial organ-at-risk contouring before qualified review and editing.
Public source Exact FDA submission Checked 2026-09-03
Required input
Adult single-energy DICOM CT images from GE or non-GE scanners.
FDA source Exact FDA submission Checked 2026-09-03
Output and human action
DICOM RT Structure Set contours and labels routed to configured review workstations.
FDA source Exact FDA submission Checked 2026-09-03
Limitations
The software is for adults and initial segmentation only. Users must verify every contour and label; average Dice varies greatly by structure and does not predict local editing burden or dosimetric acceptability by itself.
FDA source Exact FDA submission Checked 2026-09-03

Regulatory identity

FDA record and catalog context

The FDA listing establishes the regulatory identity. It does not by itself establish local workflow fit, pricing, security, or performance in your environment.

FDA submission
K230082
FDA source Exact FDA submission Checked 2026-08-31
Modality
Radiotherapy Planning Imaging
FDA source Exact FDA submission Checked 2026-09-03
Anatomy
Multi-region
FDA source Exact FDA submission Checked 2026-09-03
Clearance type
510(k)
FDA source Exact FDA submission Checked 2026-08-31
Decision date
2023-05-04
FDA source Exact FDA submission Checked 2026-08-31
FDA status
FDA-cleared
FDA source Exact FDA submission Checked 2026-08-31

Implementation

Questions for IT, informatics, and operations

Use these fields to structure a vendor demo, security review, and implementation estimate.

Integration
Server-based automated DICOM CT ingestion and RTSS export to predefined destinations; test scanner protocols, contrast, slice spacing, structure naming, routing, TPS import, and failure quarantine.
Public source Exact FDA submission Checked 2026-09-03
Deployment and data flow
Server-based software integrated with qualified GE CT and radiotherapy review workflows; exact supported infrastructure and service boundaries require vendor confirmation.
Public source Exact FDA submission Checked 2026-09-03
Security and privacy
Vendor documentation pending
Vendor confirmation Exact FDA submission Checked 2026-09-03

Public materials reviewed for this batch do not establish the complete security posture for the exact release. Request data-flow, identity, audit, encryption, retention, support-access, vulnerability-management, backup, and incident-response evidence.

Training and support
Only dosimetrists, medical physicists, radiation oncologists, and qualified radiotherapy practitioners. should operate or interpret the system. Local onboarding should cover the exact release, supported population, limitations, failure examples, escalation, downtime, and competency documentation.
FDA source Exact FDA submission Checked 2026-09-03
Monitoring and change control
Monitor by exact hardware, software and model version, patient population, anatomy, protocol, operator, input quality, output edits or overrides, failed studies, repeat acquisition or rework, downstream impact, downtime, incidents, recalls, and updates. Track the limits that matter for this product: The software is for adults and initial segmentation only. Users must verify every contour and label; average Dice varies greatly by structure and does not predict local editing burden or dosimetric acceptability by itself.
Public source Exact FDA submission Checked 2026-09-03

Release and scope verification

Map the installed model, build, options, population, anatomy, inputs, and outputs to K230082 before acceptance.

exact submission · checked 2026-09-03

Later product-family features are not evidence that this cleared release includes them.

Source

Interoperability acceptance

Server-based automated DICOM CT ingestion and RTSS export to predefined destinations; test scanner protocols, contrast, slice spacing, structure naming, routing, TPS import, and failure quarantine.

exact submission · checked 2026-09-03

Test representative identifiers, orientation, geometry, measurements, units, routing, failure handling, and round trips locally.

SourceSource

Clinical acceptance

Use a signed local test set for the supported population and workflow. Include common and difficult cases, exclusions, acquisition variation, failed or low-quality inputs, reviewer overrides, and downstream effects. The software is for adults and initial segmentation only. Users must verify every contour and label; average Dice varies greatly by structure and does not predict local editing burden or dosimetric acceptability by itself.

exact submission · checked 2026-09-03

Define stop-use, escalation, rollback, and revalidation triggers before production use.

Source

Economic evaluation

Model capital and recurring cost against net staff time after review, throughput, repeat imaging or rework, room time, consumables, infrastructure, training, service, downtime, upgrade, and replacement costs.

product family · checked 2026-09-03

Vendor productivity claims are scenario inputs, not guaranteed local savings.

Source

Lifecycle monitoring

Monitor by exact hardware, software and model version, patient population, anatomy, protocol, operator, input quality, output edits or overrides, failed studies, repeat acquisition or rework, downstream impact, downtime, incidents, recalls, and updates. Track the limits that matter for this product: The software is for adults and initial segmentation only. Users must verify every contour and label; average Dice varies greatly by structure and does not predict local editing burden or dosimetric acceptability by itself.

exact submission · checked 2026-09-03

Revalidate after material software, hardware, model, protocol, interface, or clinical-policy changes.

SourceSource

Evidence

Performance evidence

Metrics are shown only when they are tied to a source, endpoint, population, and tested product version.

Evidence summary
K230082 reports training on 911 CT examinations and independent testing on 302 exams from nine global sources, producing 2,552 expert-referenced contours. Three radiotherapy practitioners established references and separately reviewed clinical adequacy; organ-level Dice results ranged widely, reinforcing the need for structure-specific acceptance.
Public source Exact FDA submission Checked 2026-09-03
Reported sensitivity
Not applicable
Not applicable Not applicable Checked 2026-09-03

Sensitivity is not the primary applicable endpoint for this acquisition, reconstruction, segmentation, planning, guidance, or quantitative-biomarker function. Use the task-specific endpoints and tested population reported below.

Reported specificity
Not applicable
Not applicable Not applicable Checked 2026-09-03

Specificity is not the primary applicable endpoint for this acquisition, reconstruction, segmentation, planning, guidance, or quantitative-biomarker function. Use the task-specific endpoints and tested population reported below.

Independent validation exams

302 exams

Organ-at-risk segmentation in Auto Segmentation.

Tested version: GE Auto Segmentation release cleared in K230082
Source

Evaluated contours

2552 contours

Comparison with expert ground truth in Auto Segmentation.

Tested version: GE Auto Segmentation release cleared in K230082
Source

Liver mean Dice

94.28 %

Liver segmentation in Auto Segmentation.

Tested version: GE Auto Segmentation release cleared in K230082
Source

Whole-brain mean Dice

98.53 %

Whole-brain segmentation in Auto Segmentation.

Tested version: GE Auto Segmentation release cleared in K230082
Source

Chiasm mean Dice

43.81 %

Chiasm segmentation in Auto Segmentation.

Tested version: GE Auto Segmentation release cleared in K230082
Source

K230082 FDA performance package for Auto Segmentation

Manufacturer independent multicenter bench validation and blinded three-reader qualitative contour-adequacy study · 302 adult radiotherapy-planning CT exams from North America, Europe, and Asia

exact submission · Tested version: GE Auto Segmentation release cleared in K230082 · n=302 · 9 sites · Sponsor or vendor study

K230082

Manufacturer evidence submitted for substantial equivalence; local clinical, technical, and workflow acceptance remains necessary.

Source

Economics and lifecycle

Budget and ongoing governance

These are common procurement questions; unknown values remain visible until a source supports them.

Pricing and total cost
Not established in reviewed sources
Reviewed sources checked Exact FDA submission Checked 2026-09-03

No public list price for the exact cleared configuration was found. Obtain a written quote separating hardware or license basis, enabled features, implementation, interfaces, infrastructure, training, service, upgrades, renewal, consumables, and exit costs.

Reimbursement and coding
No separate named-product Medicare payment was identified in the reviewed CMS radiation-oncology material. Evaluate net planning or delivery value within the radiotherapy service line.
Public source Not applicable Checked 2026-09-03

No product-specific code

No separate named-product payment identified in reviewed public material

Centers for Medicare & Medicaid Services · United States

Build a local total-cost and value model; do not infer reimbursement from FDA clearance.

Source

Safety and lifecycle

Postmarket record

Recall and adverse-event records are shown only after product matching. Adverse-event reports do not establish incidence or causality.

Postmarket safety review
The 2026-09-02 openFDA device-recall snapshot contained no record matched to K230082 by exact submission identifier.
Public source Exact FDA submission Checked 2026-09-03

A zero-result exact-identifier search does not prove that no recall, correction, adverse-event report, or product-family safety signal exists.

Buyer worksheet

Open questions to take to the vendor

Open evaluation checklist

Research record

What has been checked

This audit distinguishes completed source review from fields that have not yet been researched.

Human reviewedStatus
2026-09-04Last searched
24Fields reviewed
8Source classes checked
20Unreviewed PubMed leads
10Unreviewed trial leads
0Unreviewed FDA recall leads

Recovered from the prior exact-submission extraction and normalized under the current provenance rules. Exact FDA materials, current product-family sources, selected clinical literature, reimbursement context, and exact-identifier recall candidates were reviewed. Pricing, contracted services, enabled options, final security architecture, and local interoperability remain organization-specific evidence. Automated exact-name discovery found 20 PubMed and 10 ClinicalTrials.gov candidate records. Candidates require human product and version matching; zero candidates is not evidence that no studies exist. Native FDA recall identifiers produced 0 postmarket candidate records; 0 have been reviewed (0 published, 0 rejected) and 0 remain unreviewed.

Candidate leads remain unpublished until a human confirms the exact product and tested version.

Source classes: fda ai list, fda decision summary, vendor product page, reimbursement policy, fda device recall, pubmed, clinical trials, openfda device recall

Sources

Source ledger

Sources accessed through 2026-09-03.

Alternatives

Related tools to compare

FDA-cleared K211881

AI Segmentation

Varian Medical Systems, Inc.

AI Segmentation version 2.0 is a cloud web application that generates initial CT organ-at-risk contours for head and neck, thorax, abdomen, and pelvis for review, editing, and...

Modality
Radiotherapy Planning Imaging
Anatomy
Multi-region
Decision
2021-09-02
Radiotherapy planningQuantification and segmentation
FDA-cleared K203469

AI Segmentation

Varian Medical Systems, Inc.

AI Segmentation version 1.0 is a static, cloud-hosted web application that generates initial CT organ-at-risk contours for head and neck, thorax, abdomen, and pelvis for review...

Modality
Radiotherapy Planning Imaging
Anatomy
Multi-region
Decision
2021-04-19
Radiotherapy planningQuantification and segmentation